Science Friday - Facial Recognition, Hummingbird Vision, Moon Lander. June 19, 2020, Part 2

Episode Date: June 19, 2020

Protests Shine Light On Facial Recognition Tech Problems Earlier this month, three major tech companies publicly distanced themselves from the facial recognition tools used by police. IBM CEO Arvind Krishna explained their company's move was because of facial recognition’s use in racial profiling and mass surveillance. Facial recognition algorithms built by companies like Amazon have been found to misidentify people of color, especially women of color, at higher rates—meaning when police use facial recognition to identify suspects who are not white, they are more likely to arrest the wrong person. Nevertheless, companies have been pitching this technology to the government. CEOs are calling for national laws to govern this technology, or programming solutions to remove the racial biases and other inequities from their code. But there are others who want to ban it entirely—and completely re-envisioning how AI is developed and used in communities. SciFri producer Christie Taylor talks to Ruha Benjamin, a sociologist, and AI researcher Deborah Raji about the relationship between AI and racial injustice, and their visions for slower, more community-oriented processes for tech and data science. Hummingbirds See Beyond The Rainbow Conventional wisdom suggests hummingbirds really like the color red—it’s the reason many commercial hummingbird feeders are made to look like a kind of red blossom. But it turns out that two items that both look “red” to humans may look very different to a hummingbird. That’s because these birds can see colors that humans cannot. Humans see colors through photoreceptors called cones, and we have three of them for red, green, and blue colors. But most birds, reptiles, and even some fish also have fourth cone that’s sensitive to UV light. That means they can see further into the spectrum than we can, and that they can see “non-spectral colors”—combinations of colors that aren’t directly adjacent on the rainbow, such as red+UV and green+UV. Mary Caswell Stoddard, an assistant professor of ecology and evolutionary biology at Princeton, set out to study whether hummingbirds actually make use of that ability in their everyday lives. Her team's research was published this week in the academic journal Proceedings of the National Academy of Sciences. A NASA Rover Is Catching A Private Ride To The Moon Last week, NASA announced that it had signed a $199.5 million contract with the private company Astrobotic to deliver NASA’s VIPER rover to the moon in 2023. The company will be responsible for the rover for getting the rover from Earth into space, up until the moment the rover rolls onto the lunar surface near the moon’s south pole. The rover is designed to explore for water and other resources—especially the large stores of water ice that scientists suspect may be frozen in lunar polar regions. Astrobotic CEO John Thornton joins Ira to talk about the challenges of building a new lunar lander, and the increasing involvement of commercial industry in the U.S. space program.   Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

This is Science Friday. I'm Ira Flato. Picture the colors of a rainbow, red, orange, yellow, so forth. We can see the rainbow colors by using three different color-sensing cones in our eyes. A dog or a cat has only two types of color-sensing cones, so their rainbow is different. Most birds and reptiles, though, have four color-sensing cones, meaning they can see colors we can't experience. Boy, are they lucky. Science Friday, produces. Alexa Lim has the rest of the story. Remember in the movie Wizard of Oz when Dorothy is flying around

in a sepia-colored tornado? But once she lands, she opens the door to the technicolor emerald city, full of color. Hummingbirds see the world differently from humans, literally, because their eyes can see a wider range of color than we do. No special glasses needed. Humans have three types of color sensing cones, for red, green, and blue light. But most birds have all those plus one more. a cone that lets them sense ultraviolet. That doesn't just give them one extra color. They can see combinations of ultraviolet plus other colors, like ultraviolet plus green.

Now researchers have found that hummingbirds can use those color combinations we can't see to distinguish and learn about the food sources they visit. Mary Castle Stoddard is an assistant professor of ecology and evolutionary biology at Princeton, and one of the authors of a study published this week in the journal Proceedings of the National Academy of Sciences. Welcome back to Science Friday, Dr. Stoddard. Thank you so much for having me. It's great to be here.

You say these birds are seeing non-spectoral colors. What does that mean? Well, that's right. We were really interested in understanding how hummingbirds perceive color. And as you mentioned, birds have the ability to detect UV wavelengths. And that's because they have a fourth color cone type that is very interesting for two reasons. The first is that it extends the spectrum of visible colors for birds. So that means that if you're a hummingbird looking up at the rainbow,

you'll see all the colors we can see, you know, red, orange, yellow, green, blue, indigo, violet, plus ultraviolet. But the second thing that the UV cone does in theory is provide birds with an extra dimension of color perception relative to humans. And that's because this UV cone type should allow birds to see a vast range of combination colors like UV green and UV red. Some of these combination colors are considered non-spectral. And when we say non-spectral, we mean a combination color that arises when the color cones that are stimulated are stimulated by light from widely separated parts of the color spectrum. So for humans, purple is the clearest example of a non-spectral color. Purple is technically not in the rainbow, but it arises when our blue and red cones, but not the green cones, are stimulated. And what's really cool about birds is that in theory, they have many different kinds of non-spectral color.

So we humans have just purple, but birds can theoretically see purple, UV red, UV green, UV yellow, and UV purple. Wow. And you said, you know, it's more than just an extended color pell. It's an extra dimension? Yes. One way we like to think about this is in terms of a color space. So we tend to represent human colors in a color space that looks like a triangle. Each vertex represents one of our three color cone types. But for birds, we often represent the colors they can see in a tetrahedron. And each of its four vertices represents one of the bird four color cone types. And so we predict that birds can see colors throughout this tetrahedral color space.

Wow. Can you give me an example of what kind of information is coded in this extra color sense then? What are they using it for? Well, it turns out that these kinds of combination colors are really common in the environments of birds. So we were able to analyze a large data set of flower and plumage feather colors. And we could predict that about 30 to 35 percent of those colors would be perceived as non-spectral by birds. But a much smaller fraction of those colors would appear as non-spectral to humans. So we think that these colors are, you know, they're out there, they matter, they're ecologically relevant for these birds.

They're using them to find food. They're sometimes using them to attract mates. For example, the male broad-tailed hummingbirds perform spectacular courtship dives. And as a male zips over the head of a female, his magenta, iridescent throat feathers are on full display. And we estimate that that magenta color would appear to birds as a non-spectral UV purple. We don't think that non-spectral colors are particularly special relative to other colors in the environment for birds. They are just part and parcel of a tetragrammatic color vision system. So these colors are interesting to us because we see them as very different,

and they haven't received much formal research attention. But what matters to a bird when it's looking at a color isn't whether it stimulates this cone or that cone, It's what does this color reveal about, mates, predators, or food? So colors are, they're kind of arbitrary. I mean, it's not like there's like a definitive color. It's just kind of how they're being interpreted in our brain. That's absolutely right. It is a sensation that arises in the brain, and in that sense, they're subjective.

It's an interaction between the outside external world and wavelengths of light that are being reflected and how our eyes and brains are interpreting those wavelengths of light. Wow, that's pretty cool. So I also want to talk about how you tested this, because I know you used wild hummingbirds. That doesn't sound like the easiest setup. Definitely not. So what did you do to test this?

Well, we trained hummingbirds to participate in color vision experiments. Each morning we got up really early. and we set up two feeders. One feeder contained sugar water and the other contained just plain water. Next to each feeder, we placed a special LED light tube, and we designed these tubes to display a whole range of bird visible colors like UV green and UV red.

The tube beside the sugar water emitted one color, while the tube next to the plain water emitted a different color. and we frequently swapped the positions of the rewarding and unrewarding tubes, so the birds couldn't just memorize the location of the reward. We also wanted to make sure that the birds weren't using smell or some other cue to find the reward. So we performed a series of control experiments. Our feeders were set up in the middle of a meadow, and this meadow was visited by dozens of hummingbirds,

So we could observe whether the birds tended to visit the rewarded or the unrewarded feeder when they came back in search of, you know, sugary snack. So over the course of several hours, if the birds could distinguish between the two colors we were testing, we saw that they tended to visit the rewarded feeder. And using this setup, we were able to show that hummingbirds can see a variety of these non-spectoral colors, such as UV red, UV green, purple, and UV yellow. All the hummingbird feeders we put out there are usually red. For some reason, we think hummingbirds love red.

Do they actually prefer that color? They do love red, but they don't have an innate preference for it. They have learned to associate red with highly rewarding flowers or sometimes highly rewarding artificial feeders. You know, most bird feeders are red. And in our experiments, we show that how many birds could indeed learn to associate a red light with a reward, but they quickly lost this preference when we rewarded another color instead. So they can rapidly unlearn this preference for red if something else is rewarded. Okay. So they're pretty trainable. We're training them.

They really are. We said that we trained them to participate in color vision experiments, but really they required very little training. They're ideal test subjects for a study like this. Right. So humans have three cones, birds have four cones. What happened to our cone? Why did it drop out of the evolutionary tree or did birds evolve more? Well, that's a great question. So birds actually didn't evolve an extra cone.

They have retained the ancient vertebrate color vision system, which had these four different color cone types. So many fish, reptiles, even dinosaurs. are or were tetragrammatic with these four color cone types. It turns out that mammals lost two of the color cone types early in their evolutionary history when they were mostly nocturnal. So we humans enjoy our very modest color vision only because old world primates re-evolved a third color cone type by mutating one of the existing cone types. So the point gear is not that birds are superior, it's that humans are so basic. You've said it, yeah, right? Well, I guess on that note, are there any animals that are less basic at that now? So that have

more than four comotypes? Well, we will be considered extremely basic when we look to other animals and the animal kingdom. So the answer to that is yes. Probably the most famous in this respect are mantis shrimps, which have up to 12 different photoreceptors for color. Butterflies can have up to nine distinct color photoreceptors. So there are certainly animals out there with really interesting color vision. I think we have a lot more work to do in our field to understand how these photoreceptors are wired and whether they all contribute to color vision in the way that we expect. But these are really big open questions in our field.

There's this entire communication system that's out there that we're not even attuned to. We can't even tap into. That's really true. I mean, these are the things I daydream about. You know, what is this color experience really like for birds? You know, is it like a quantum leap in color experience? You know, the same way that color TV is just so much better than black and white TV. I mean, the truth is that that we don't know.

And we cannot say, you know, we don't really know what a UV red color looks like to birds. Is it a mixture of those two colors or is it a sublimely new different color? We can only speculate. And I think that's what makes studying bird perception, you know, both so tricky and so fun. Well, thanks so much for joining us. Thank you so much for happy me. Mary Castle Stoddard is an assistant professor of ecology and evolutionary biology at Princeton. For Science Friday, I'm Alexa Lim.

We're going to take a break, and when we come back, why some companies are rethinking facial recognition and why some researchers want to ban it entirely and not just for police. Story coming up, stay with us. This is Science Friday. I'm Ira Plato. It's been a big couple of weeks for facial recognition technology. IBM said it would stop using it entirely. Meanwhile, Microsoft and Amazon have paused allowing police to use their facial recognition technologies, at least until there is a national law to ensure its use doesn't perpetuate racial inequities.

Sci-fi producer Christy Taylor talked to two experts who have a different idea, ban facial recognition entirely, and rethink how we develop new AI technology, while we're at it. So is this a national moment for facial recognition, too? If so, it's been building for a few years. San Francisco banned facial recognition use by police and other government agencies last year. And research revealing huge disparities

in how accurate facial recognition is. That's been around since MIT research in 2018, which found that facial recognition is most accurate if you're a white man, and least if you're black, a woman, or both. We've talked about technological solutions, for the biases that can be built into AI before. But like the national conversation around policing,

there are people who don't just want to reform AI but actually stop investing in technologies that are too harmful to reform. Here to talk about why that is, Dr. Ruha Benjamin, professor of African American studies at Princeton University, and author of Race After Technology, Abolitionist Tools for the New Jim Code. Welcome back to Science Friday, Dr. Benjamin.

I'm thrilled to be here. Thanks for inviting me. And also here we have Deborah Raji, a technology fellow at the AI Now Institute at New York University. Thank you for joining us, Deb. Lovely to be here. Thank you for inviting me. Yeah, you're welcome. Ruha, I'm going to start with you because in the last week, we've seen IBM say they were divesting entirely from facial recognition. And Amazon and Microsoft say they would stop selling their products to police, at least for a while.

IBM CEO, Arvin Krishna, condemned the use of facial recognition software in racial profiling and mass surveillance. Why are we seeing this now? I mean, I think it speaks to the power of protests, the power of public condemnation against policing in general, the police abuses we've seen. And companies far and wide are trying to distance themselves from what people are rightly criticizing. And so in addition to these tech companies, we've seen everything from, you know, Hollywood movies and shows, cops being taken off the air to NASCAR banning the Confederate flag. So I think it's part of a spectrum in which people are understanding this cultural shift is not going anywhere. So they have to respond in kind. Deborah, how widespread is police use of facial recognition technology?

Will these moves make any kind of a difference? Yeah, I want to just emphasize that since summer 2018, there's been so much attempt and so much effort from ACLU, but also a lot of other advocates and technology advocates to try to expose and discuss. the reality of the use of facial recognition by police. So ACLU since summer of 2018 and likely even before that has been sort of investigating Amazon's, Amazon in particular, is interaction and attempt to sell that technology to police departments. Amazon on their website advertises at least one police department, which is Origin County that we know of. But there's also sort of reports from different groups, including workers at Amazon,

on claiming there to be more clients and ACLU themselves identifying sort of pitch decks to departments in Orlando and other regions. And even if just the one department that is advertised on their website, even if that one department is making use of facial recognition, that's still affecting thousands and thousands of people. So it really is sort of an impactful decision for them to pull away from the technology, especially as this more nuanced conversation around its more widespread spread use is happening. I'm very grateful to sort of see the conversation get to the point where they understand that this technology cannot continue to be sold while the policy conversation is happening. So, yeah, I do think that it is an impactful decision and it will sort of be directly correlated

to protecting, you know, several people to the order of thousands of people from harm. Why policing? Why is that where so much of the worry about facial recognition is concentrated, Deb? It's not just policing to clarify, right? There's use of facial recognition. and certain hiring tools like we saw with Higher View, there's a lot of interest from the Department of Homeland Security and using facial recognition as part of the immigration process. It very much is a part of the fabric of American life in different ways. I think policing is an alarming one,

and facial recognition is this technology that is very easily manipulated and very centralized. You have a lot of identifiable biometric information about a lot of people, it requires a certain amount of compute to like create the model, to put that in the hands of an authority figure that, you know, we're beginning to question and we're beginning to distrust, I think, is really at the heart of a lot of this conversation that we see today around like, do we trust the police with this technology that can be so easily weaponized? And then also, you know, historically looking at why facial recognition was sort of encouraged to be developed in the first place, a lot of the early investors in the technology, like the National Institute of standards was sort of the first group to really build a lot of these big face data sets to sort of kickstart the industry in the U.S. And a lot of their early funders were coming from intelligence agencies and sort of with an eye towards law enforcement thinking about like mugshots. You know, we have a lot of face data connected to the law enforcement paradigm in

the U.S. So it's very easy to use facial recognition for that purpose. And in the last couple weeks, we've sort of seen how just because it's very easy to use it and it's sort of this very important tool for them does not necessarily mean that they are the right people to entrust with this tool. And I would say one of the dangers is that we sort of take this win and then become complacent. Because the line between law enforcement and so many of other our institutions is very porous. And so, for example, when schools use facial recognition, UCLA was about to implement a facial recognition system to look at people as they're coming on campus

to determine if someone was an actual student, faculty, or staff. And when fight for the future, a digital rights nonprofit analyzed the system, they found that it came back with 58 false positive matches that were largely students of color. So you can imagine that a black student is walking across campus falsely flagged as an intruder, and the police are called,

and what will happen in that instance, and we see what's happening on the street when the police arrive and decide that a black person is to be targeted. And so in this case, when educational institutions, private sector, companies, all kinds of public spaces employ this, the police are right behind them.

And so it's not enough just to ban it on the part of police when other institutions and entities use a tool that's not only scientifically falsie, but also one that has deep, racist roots. I want to talk more about those inaccuracies that you just referred to, because Deb, I know you were a co-author on some of the research that uncovered those disproportionate inaccuracies. Tell us more about that. Yeah, I was sort of involved in this project led by Joy Blu and Weeney at the Al arithmetic Justice League, and at the time she was a

grad student at the MIT Media Lab. And she sort of was able to identify the fact that in computer vision and research, the way that these models were sort of trained and evaluated were on these test images that didn't necessarily represent the full scope of the populations that they were being implemented on. This reflects sort of my early experience where I was working on an applied machine learning team and I was noticing that a lot of the data sets that I had to work with did not include anyone that looked like me. So there were not a lot of darker skin people. There were not a lot of even women. So Joy really led. the effort to really ask the question of, you know, what would happen if we created an evaluation

test set that actually represented the full range of skin types that we have. So, you know, darker skin and lighter skin and was balanced with respect to those different skin types, but also gender and looking at the intersection, the performance at the intersection of these axes. So she created this project called Gender Shades that was really that first critical evaluation of, like, how does this deployed product? And this is something that I like to remind people as like, At the time when we audit these systems, you know, they're already out there in the world. You know, they're products that are already sold, already integrated into applications, you know, who knows where. So we said, you know, looking at these products that, you know, have already been deemed good enough to throw it into the world,

how well does it actually work for these different subgroups? And what we found was that there was almost a 30% disparity between the darker female subgroup and the lighter male subgroup. And there was a first round of audits on IBM Microsoft and Face++, which, which is a Chinese facial recognition company. There was a very public response to that initial audit, and we were thinking, oh, maybe this represents a shift with respect to the industry. So we did a follow-up audit to sort of see how the companies that we had audited responded, but also some of these other companies, including Amazon.

And what we found was that, you know, even after being witnessed to other companies getting audited and understanding that there was a racial bias issue that existed within the facial recognition space, Amazon still demonstrated disparities of, you know, over 20%, 30% between the darker female subgroup and the lighter male subgroup. You know, I remember when I first began to just holistically notice, like, oh, there's not a lot of black people in these datasets while I was, when I started working. When I first noticed these things, I remember trying to have conversations with my manager at the time. And he was kind of like, it's so hard to collect data at all. Why would we think about representation?

Why would we think about diversity? It was such an ingrained attitude at the time to ignore the problem because it was just too hard. I think now we're at a point where there's like now this acknowledgement that like, no, we should create representative data sets, but also this is a great starting point to really questioning the functionality of these technologies. Does facial recognition really work if they were already deploying a version of this technology that did not work for black people or for, you know, darker skin people? you know, does it really work if, you know, when they try to attempt to diversify the datasets, there's, you know, privacy violations that are discovered. Does it really work if it's so easily weaponized by institutions that we no longer trust, you know? So a lot of these questions really just spewed out of that project. So I'm grateful to have participated in that.

And one of the things I really appreciate about your work, Deb, about Joy's work, is that after the initial sort of technical faultiness of these systems were revealed. The goal wasn't simply to perfect the system to make them more accurate at detecting people when the actual mechanisms of identifying people are themselves unjust and unethical. And so the whole project is not simply about honing these tools, but to actually use the faultiness to pose these larger questions about whether we want these at all. And so the goal, is not simply better tools and more accurate tools when that would only likely just lead for more a more honed injustice in the process, better able at identifying the most vulnerable in our

communities. And so I just wanted to add that is that we're not questioning simply the scientific merit of these systems, but they're ethical and their political merit. Just a quick reminder that this is Science Friday. I'm Christy Taylor. Talking about rethinking AI, with Dr. Ruhah Benjamin and Deborah Raji. Ruhan, you mentioned this idea of a more honed injustice. What kinds of harms exactly do you see from something like that? I think of facial recognition as part of a whole family of technologies, automated AI-based technologies that have been rolled out under the guise of neutrality,

what I call the new gym code. We see it when it comes to administering public benefits. We see it in health care. We see it in our prison system. So, for example, in the midst of the pandemic, there's been a outcry about the overcrowding in our jails and prisons. And so one of the responses has been a technical fix. Let's use a risk assessment tool, one called Pattern, to decide who is the least risky to release so that we can deal with this overcrowding. And pattern, this risk assessment tool, was found, first of all, is scientifically unverified.

And then those who've audited it have found that 7% of black men were classified as, minimum risk and able to be released compared to 30% of white men. And there was biases associated with homelessness and mental illness. And so here's an example of a technical fix that's posited as a solution for some pandemic-related crisis, in this case, overcrowding in prisons, that has this racial bias baked into it. And this is just one of many and one of the most recent in which we see that the turn to automation and automated decision systems, very likely the default settings will lead to an exacerbation of inequalities, existing inequalities, and hiding these inequalities behind a veneer of neutrality and objectivity that makes it even

harder to question and hold accountable. If it was a biased judge sitting up there or a biased prosecutor, at least you could point to the person. But in this case, people point to a screen and say, this thing can't make decisions. This thing doesn't have a grudge against people or a hatred against people. And yet baked into it are patterns of profiling and discrimination that then get hidden behind a statistic or a score. And so it cuts across almost every institution, education decisions, health care decisions, public benefits. It's penetrated every area of our lives.

And many people don't even realize that very consequential, you know, decisions in their lives are being made by automated systems that are exacerbating inequalities. Deb, do you have any other examples you would point to? I like to sort of remind people that a face is sort of the equivalent of a fingerprint with respect to its role and its status as an identifiable biometric. Like, we do not upload pictures of our fingerprint to the internet. We're very careful about that data. and we should be just as careful about face data. So because our faces are these identifiable things and we upload them so freely,

a lot of companies, including Clareview, have sort of found a lot of success in terms of this idea of digital surveillance and just being able to match different profiles online using the photos that people post, being able to track people for the sake of whatever authority figure. So there's stories of, you know, ICE tracking particular suspects

using their information of their different social media outlets using their face data. So I see that being an alarming example of the use of facial recognition outside of sort of the CCTV camera and identifying your face as you're walking down the street. There's a lot of online information that facial recognition makes its way through and organizes for other people. And then the other sort of example that I think of often is the case of the Atlantic Towers apartment with the Brooklyn tenants. It's a case I reflect on a lot.

So to just give a short recap, you know, there's these tenants in this rent-controlled building in Brooklyn, and they find out that their landlord who sort of had a history of racial bias and has incentive to sort of evict tenants, wants to install a facial recognition system. And the tenants are against the facial recognition system. And it's sort of a recorded case that on the surface feels like, oh, you know, it's a majority black. community, maybe they're worried about accuracy or maybe they're worried about privacy because the data is not encrypted and they're not sure where the data's going to go. When I started having conversations with the tenants, I realized their fear was really the way that the landlord

could very easily weaponize that technology to monitor them, to monitor them coming in and out. It was like a threat to their safety in that sense of the technology being sort of weaponized as a method of controlling that environment and really, putting them at risk with respect to this authority figure that they couldn't trust. So, yeah, those are sort of, I guess, some of the cases that I reflect on a lot. We have to take a break, but when we come back, more on reimagining our relationship with artificial intelligence. Plus, NASA has signed a contract with a private company to deliver a rover to the south

pole of the moon to look for water. You're going to want to hear that. So stay with us. This is Science Friday. I'm Ira Flato. In case you just joined us, we're talking about facial recognition, artificial intelligence, and developing technology without harmful biases. Producer Christy Taylor spoke with Dr. Ruha Benjamin, professor of African American Studies at Princeton University, and Deborah Raji, a technology fellow for the AI Now Institute at New York University. Ruha, I want to go back to police because I saw on your Twitter feed the other day, you said, with technology, we can police without the police. What did you mean by that? And is that a good thing?

Yeah, absolutely. And what's interesting is that people interpreted, some people obviously who were commenting seem to interpret it as a kind of encouragement of policing without the police when it was a critique. And in fact, Deb's last example is a prime example because in this case, you have a private housing developer implementing facial. recognition and exercising, you know, forms of containment and control and surveillance without someone standing there with a badge and a uniform checking people out. So that's an excellent example of the same logics and practices of policing that keep a watch on people, that profile people, all of those things that can be exercised without the institution of the police. So my concern is now in this moment when we're focused on defunding the police, that we stop looking at the ways that racism is mercurial. It takes different forms. And as soon as you may lessen the number

of policing in your town or city, but other institutions take up the work of policing by, for example, implementing facial recognition or other types of surveillance tools. the idea of abolition that is becoming more mainstream in this moment, it has a dual meaning that is to destroy and to grow in the etymology of this word, abelare. And so we have to think about what we want to get rid of, but also what we want to grow, because if we're not growing alternative institutions, practices, and ways of life, then that old institution is going to take on a new form. It's going to shape shift. It's going to be exercised through various kinds of more invisible forms of policing that, again, will be hidden behind a veneer of neutrality. What about the conversation about

abolishing versus reforming the police? Is there a parallel in the AI world? You know, the, we need better training data versus this technology is too dangerous to exist. Yeah, I think the conversation, one form that the conversation takes is to be wary and to push back against tech fixes for social problems. Even now, there are people who are positing various kinds of apps to help deal with police violence, right? And so finding a technical fix that sort of papers over the deep roots of a problem means that you might deal with certain symptoms of an issue, but the underlying issue will come up in a different form if we don't keep our eye on that. And so just thinking about police reform first, we know that in Minneapolis, that point,

police department had implemented so many of the different reforms that many people call for. They implemented implicit bias training, CAMs, community engagement, and mindfulness training, all of the things we can call for. And yet he still died. He was still murdered. And so again, it underlies that simply tweaking an institution that was born out of slave patrols, that was born out of a desire to contain people, is never going to get us where we want to go. So again, with tech fixes, we have to think about what is the underlying cause. For me, the one area where I think AI and data collection and all has a role to play is when we flip the lens back onto those with power and actually use it to expose issues, not necessarily try to fix them. So for example, when it comes to

housing discrimination, there's a wonderful initiative called the Anti-Eviction Mapping Project, which turns the lens onto landlords rather than tenants. And it looks at the practice of evictions. It looks at different cities and it finds patterns of evictions. It lists worst evictors in different cities. And now during COVID, it's paying attention to the way that as the various kinds of eviction moratorium are running out, what are the underlying issues that are causing this? So when we turn the digital lens onto those who monopolize power and resources and use it to expose problems, I think it has a role to play. But again, we have to think about what question are we posing that we want the technology to answer? And there you're going to find the seeds of either the ability to subvert power or really

reinforce the existing power relations. Yeah, I think reform, especially in the case of, you know, post-Ferguson, a lot of the reform measurements that were proposed required investment in the police. The proposal for, you know, cameras, for example, like that actually gave more money to the police departments to invest in that technology. And we see that sometimes with response to sort of revelations of bias in the technology, people will be like, oh, let's just like invest more in facial recognition to address the bias. And I think putting that on its head and asking to defund that technology, defund the police or ban it. And I think that's like such an incredible sort of counter narrative to say like actually like rather than investing more energy into this space,

why don't we actually just take a step back and completely reinvent the scope of solutions we're thinking of for the issues that are the real underlying issues that we're trying to address here. A lot of AI researchers are definitely going through a phase of understanding that maybe we don't want to invest more of our resources and time and effort into improving facial recognition. Maybe there's just so many dimensions of concerns here that we need to just like take a step back from this field and let it go. And also, you know, really advocate for the big tech companies and also the smaller tech. companies to stop the sale of this technology and really restrict it to use in a significant way by advocating for that kind of policy. Well, it was really wonderful to talk to you both today. Thank you so much for your time. Thanks for having. Thank you so much for having us. Yeah, this is awesome. Dr. Ruha Benjamin, professor of African American studies at Princeton University, an author most

recently of Race After Technology, Abolitionist Tools for the New Jim Code, and Deborah Raji, a technology fellow for the AI Now Institute at New York University. And just a quick note, this interview was actually much longer. Keep an eye out on the Science Friday podcast feed for Ruha and Deborah's policy wish list and vision for more ethical tech. Coming soon, wherever you get your podcasts. Plus, you can learn more about their work and more about calls to ban facial recognition technology from communities like Detroit on our website, ScienceFriiday.com slash community AI. For Science Friday, I'm Christy Taylor. NASA wants to go prospecting for water on the moon.

And last week, the space agency announced that it had awarded a contract to a private company to deliver a NASA rover called Viper to search the South Pole of the Moon for water ice. The price tag, just under $200 million. The company Astrobotic aims to provide end-to-end delivery of a golf cart-sized rover in 2023. It's part of a program called the NASA Commer. commercial lunar payload services program. Private companies delivering stuff to the moon, and Astrobotic has plenty of plans to deliver all kinds of stuff to the lunar surface.

Joining me now to talk about the mission and the commercial space industry is John Thornton, CEO of Astrobotic, based in Pittsburgh, Pennsylvania. Welcome to Science Friday. Oh, thank you, I, are. Good to be here. John, your contract with NASA involves end-to-end delivery of their rover viper. What do you mean by end-to-end delivery? That's right. For our lander delivery service to go to the moon, what we do is we provide literally end-to-end.

And in our case, it's kind of like the shipping model, like a DHL delivery truck. We accept a package from NASA. In this case, it's a large rover. And then we deliver that all the way to the surface of the moon. So the way that works on our side behind the scenes is we go and buy a launch vehicle. We build a spacecraft, in this case a lander that goes on top of that launch vehicle. and then that lander gets launched up into space thrown towards the moon, and then our job is to do trajectory correction maneuvers along the way to make sure that we're lined up.

We slow down, we get to Earth lunar orbit, and then we just slow down again to lower that orbit and slow down one last time for terminal descent for a soft landing down on the surface. And at that point, we operate payloads and deploy the payloads. So your job is to bring payloads to the moon without having to actually use NASA in between except for maybe renting their launch site? That's correct. And even the launch site, we don't need to rent from NASA.

The commercial launch providers already have their own launch pads. So it's completely without NASA, but of course it is for NASA and funded by NASA. So you go out and you hire a rocket booster maker. You send out bids. They give you the booster. You put your own payload in it and off it goes once it leaves the booster. It's your baby to fly to the moon. That's right.

So from there, we are controlling our lander and the crews and ultimately the landing and also surface operations, all from our mission control in the north side in Pittsburgh. And what's the business model there? You say you want to bring stuff to the moon, stuff from everybody. That's right. Our goal is the company is to make the moon accessible to the world. And for us, that means not just space agencies like NASA and even other space agencies we have on our first mission, like the Mexican space agency. we really want to make it also possible for universities and even individuals to fly things up to the surface of the moon.

We have a program called DHS Moonbox, for example, that does just that. For a few hundred dollars, folks can send small memento's up to the surface of the moons. We have photos of families. We have inscriptions on metal. We have folks that are even sent up some hair from a family pet that had passed. So really, whatever story individually that you want to have connected to the moon for all time can now be done. We can bring momentos up there and store them in a time capsule. This program has been quietly flying under the radar.

I think many of us have never heard about this before. Yeah, it's a funny secret. It's not intentional. We have been trying to get out there as much as we can. We recently bought a building in Pittsburgh, and we've got about 74 employees, and we're going to be adding to that. So we hope that Pittsburgh and the whole region will follow along with our return to the moon, leading America back to the moon, in fact,

and these first commercial deliveries are going to be major, major historic moments. And what's going to be extremely exciting is all this picture and video coming back from the surface of the moon will all be processed in Pittsburgh and then put out for folks around the world to be watching.

Let's talk a bit about the details of the mission to extract water from the polar regions of the moon. Why is that so important? Yeah, NASA is sending a rover called Viper to the polar. the moon, and the goal there is to extract water. So what it wants to do is go to the spots where permanently shadowed areas that are super, super cold are likely sources of water. And the idea with the Viper rover is to send a drill down into the subsurface parts of these permanently shadowed areas,

find out if there's ice stored there, what quantity, what composition, what's it attached to? to ultimately eventually send subsequent missions to potentially extract that water and turn it into rocket fuel. Water, you can split, oxygen and hydrogen. That's the same fuel that powered the shuttle. So if we can harness rocket fuel from the moon, that would be the first time that we're using resources from another planetary body for our own purposes. And that is a huge advancement for our species in space and ultimately the development and exploration of space. And so you have a drill on board or NASA has a drill on board that I understand that you helped pioneer back when you were at Carnegie Mellon? That's right. Yeah, it's really fun, full circle story. So 13 years ago when we started, before we started the company, in fact, I built a rover with Red Whitaker, Carnegie Mellon, where it was a rover that was a concept rover to put a drill on a rover, drive it to potentially drill for water at the poles of the moon.

13 years later, that same drill and the same technology evolved, of course, is now going to be sent to the surface of the moon. And it's really a great honor 13 years later to be a part of that and amazing that ASSA gave us $200 million to make that happen. And we're just so excited, thrilled, and honored to be a part of that. And personally, it's a big one for me. Just a quick note, I'm Ira Flato, and this is Science Friday from WNYC Studios. talking with John Thornton, CEO of Astrobotic, a commercial space services company based in Pittsburgh, Pennsylvania, which is seeking to send stuff to the moon. $200 million is, for those of us who follow these kinds of things, pretty cheap for a mission to the moon. How are you able to pull that off?

That's right. This is a commercial, affordable access to the moon is our goal. $200 million is relatively cheap for a science mission and a planetary mission. this type. The way we're able to do that is we use commercial launch first, and that has a lot of efficiencies in it. And then for us as a company, we are built as a commercial company first. So that means efficiencies from costs looking at acquisitions and what's a buy in-house. We have a risk profile that's associated with a commercial approach. We aren't trying to innovate technology, just to innovate technology. We're really trying to find the lowest cost, highest performance

sweet spot for operating our business. And that provides a new paradigm that is quite cost competitive. And that's really important for NASA because that frees up their resources to do the bigger and harder things in space. So you have all these missions going before you send NASA's Viper to the moon's surface? That's correct. Our first mission flies next year. It's a lander called Paragrin. It delivers about 90 kilograms of payload to the surface of the moon. And that's a mixture of science instruments from NASA and commercial payloads from around the world as well as other space agencies. That mission alone will probably triple the number of nations that have ever operated payloads on the surface of the moon. And then the following year in 2022, we actually have a rover going on

someone else's lander that will fly up to the surface. That's a rover to demonstrate autonomy, and that's an important aspect for the development of moon and other planetary destinations. And then, of course, this new mission that we just got last week is going to go at the end of 2023. So what is this rover going to do, this other mission? So the mission in between our lander programs, the 2022 mission, this is an autonomous rover called Moon Ranger. And the idea there is to demonstrate the autonomy on the moon so that we can have the capability to drive around the moon without direct communication back to Earth. And that's very important because there are communication blind spots at the moon, especially at the polls, where potentially you could drop signal from Earth and you need to be able to reestablish that communication back to Earth. And who is picking up the tab on that one?

So that one is a NASA-funded payload that we are developing. And then one of our competitors is flying that to the surface of the moon on another commercial lunar payload services contract. Do you foresee the price coming down even cheaper than $200 million as you ramp up production? We hope so. A big driver for our cost is launch costs, so we hope to see launch costs continue to come down as well. So I think in concert, we do hope to make it even more and more affordable over time. Well, we wish you good luck, and we'll be watching closely. Thank you for taking time to be with us.

Well, thank you, Ira, and thanks for telling our story and let me be on today. Appreciate it. You're welcome. John Thornton is CEO of Astrobotic, a commercial space services company based in Pittsburgh, Pennsylvania. And that's it for this hour. If you missed any part of this program or you'd like to hear it again, subscribe to our podcasts or ask your smart speaker to play Science Friday. And our question this week on the Science Friday Vox Pop app, have you started socializing a little bit more post-lockdown? How do you decide what's safe and what's not? And how are you protecting yourself? Really good, interesting questions.

We want to know. We want to hear your answers on our Science Friday Vox Pop app wherever you get your apps. You can also say hi to us on social media, Facebook, Twitter, Instagram. You can always email us the classic way, sci-fri at science friday.com. Send us feedback.

Tell us what you would like us to cover. We'd like to hear from you. Have a great weekend. I'm Ira Flato.